html5-img
1 / 40

Topic-Sensitive SourceRank: Agreement Based Source Selection for the Multi-Topic Deep Web Integration

Topic-Sensitive SourceRank: Agreement Based Source Selection for the Multi-Topic Deep Web Integration. Manishkumar Jha Raju Balakrishnan Subbarao Kambhampati. Deep Web Integration Scenario. Millions of sources containing structured tuples. Autonomous Uncontrolled collection.

kalani
Download Presentation

Topic-Sensitive SourceRank: Agreement Based Source Selection for the Multi-Topic Deep Web Integration

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Topic-Sensitive SourceRank: Agreement Based Source Selection for the Multi-Topic Deep Web Integration ManishkumarJha RajuBalakrishnan SubbaraoKambhampati

  2. Deep Web Integration Scenario Millions of sources containing structured tuples Autonomous Uncontrolled collection Contains information spanning multiple topics Mediator Access is limited to query-forms ←answer tuples ←query query→ answer tuples→ answer tuples→ ←query answertuples→ ←answer tuples query→ ←query Web DB Web DB Web DB Web DB Web DB Web DB Web DB Deep Web

  3. Source quality and SourceRank Source quality SourceRank SourceRank[1] provides a measure for assessing source quality based on source trustworthiness and result importance • Deep-Web is • Uncontrolled • Uncurated • Adversarial • Source quality is a major issue over deep-web [1] SourceRank:Relevance and Trust Assessment for Deep Web Sources Based on Inter-Source Agreement, WWW, 2011

  4. Why Another Ranking? Rankings are oblivious to result Importance & Trustworthiness Example Query: “Godfather Trilogy” on Google Base • Trustworthiness (bait and switch) • The titles and cover image match exactly. • Prices are low. Amazing deal! • But when you proceed towards check out you realize that the product is a different one! (or when you open the mail package, if you are really unlucky) Importance: Searching for titles matching with the query. None of the results are the classic Godfather

  5. SourceRank Computation • Assesses source quality based on trustworthiness and result importance • Introduces a domain-agnostic agreement-based technique for implicitly creating an endorsement structure between deep-web sources • Agreement of answer sets returned in response to same queries manifests as a form of implicit endorsement

  6. Method: Sampling based Agreement where induces the smoothing links to account for the unseen samples. R1, R2 are the result sets of S1, S2 . • Agreement is computed using key word queries. • Partial titles of movies/books are used as queries. • Mean agreement over all the queries are used as the final agreement. Link semantics from Si to Sj with weight w: Si acknowledges w fraction of tuples in Sj. Since weight is the fraction, links are unsymmetrical. SourceRank defined as the Stationary Visit Probability on the Agreement Graph

  7. Computing Agreement is Hard Computing semantic agreement between two records is the record linkage problem, and is known to be hard. • Example “Godfather” tuples from two web sources. Note that titles and castings are denoted differently. Semantically same entities may be represented syntactically differently by two databases (non-common domains).

  8. Detecting Source Collusion The sources may copy data from each other, or make mirrors, boosting SourceRank of the group. Observation 1: Even non-colluding sources in the same domain may contain same data. e.g. Movie databases may contain all Hollywood movies. Observation 2: Top-k answers of even non-colluding sources may be similar. e.g. Answers to query “Godfather” may contain all the three movies in the Godfather trilogy.

  9. Factal: Search based on SourceRank ”I personally ran a handful of test queries this way and gotmuch better results [than Google Products] results using Factal” --- Anonymous WWW’11 Reviewer. http://factal.eas.asu.edu [WWW 2010 Best Poster; WWW 2011]

  10. SourceRank is Query Independent • SourceRank computes a single measure of importance for each source, independent of the query… • Large source that has high quality for one topic will also be considered to be high quality for topic B • Ideally, the importance of the source should depend on the query • But, too many queries.. • ..and too costly to compute query specific quality at run time..

  11. ..But, Sources May Straddle Topics Mediator ←answer tuples ` ←query query→ answer tuples→ answer tuples→ ←query answertuples→ ←answer tuples Movie query→ ←query Web DB Web DB Web DB Web DB Web DB Web DB Web DB Deep Web Camera Books Music

  12. … And Source quality is topic-sensitive • Sources might have data corresponding to multiple topics. Importance may vary across topics • SourceRank will fail to capture this fact • Issues were noted for surface-web[2]. But are much more critical for deep-web as sources are even more likely to cross topics book movie [2] Topic-sensitive PageRank, WWW, 2002

  13. SourceRank is Query Independent • SourceRank computes a single measure of importance for each source, independent of the query… • Large source that has high quality for one topic will also be considered to be high quality for topic B • Ideally, the importance of the source should depend on the query • But, too many queries.. • ..and too costly to compute query specific quality at run time..

  14. This Paper: Topic sensitive-SourceRank • Compute multiple topic-sensitive SourceRanks • Source quality is a vector in the topic space • Query itself is a vector in the topic space • At query-time, using query-topic, combine these rankings into composite importance ranking • Challenges • Computing topic-sensitive SourceRanks • Identifying query-topic • Combining topic-sensitive SourceRanks

  15. Computing topic-specific agreement • For a deep-web source, its SourceRank score for topic will depend on the answers to queries of same topic • Topic-specific sampling queries will result in an endorsement structure biased towards the same topic • Topic Specific Source Ranks are stationary visit prob on the topic-specific agreement graphs Movies Books

  16. Computing Topic-specific SourceRanks • Partial topic-specific sampling queries are used for obtaining source crawls • Biased agreement graphs are computed using topic-specifc source crawls • Performing a weighted random walk on the biased agreement graphs would result in topic-specific SourceRanks, TSR’s

  17. Topic-specific sampling queries • Publicly available online directories such as ODP, Yahoo Directory provide hand-constructed topic hierarchies • Directories are a good source for obtaining topic-specific sampling queries

  18. Query Processing • Computing query-topic • Computing query-topic sensitive importance scores

  19. Computing query-topic • Query-topic • Likelihood of the query belonging to topics • Soft classification problem topic query-topic For Query=“godfather”

  20. Computing query-topic – Training Data • Training data • Description of topics • Topic-specific source crawls act as topic descriptions • Bag of words model

  21. Computing query-topic – Classifier • Classifier • Naïve Bayes Classifier (NBC) with parameters set to maximum likelihood estimates • NBC uses topic-description to estimate topic probability conditioned on query q where qjis the jth term of query q

  22. Computing query-topic sensitive importance scores • Topic-specific SourceRanks are linearly combined, weighted based on query-topic, to form a single composite importance ranking where TSRki is the topic-specific SourceRank score of source sk for topic ci

  23. Agenda • SourceRank • Topic-sensitive SourceRank • Experimental setup • Results • Conclusion

  24. Four-topic deep-web environment

  25. Deep-web sources • Collected via Google Base • 1440 sources: 276 camera, 556 book, 572 movie and 281 music sources

  26. Sampling queries • Used 200 random titles or names for each topic

  27. Test queries • Mix of queries from all four topics • Generated by randomly removing words from titles or names with 0.5 probability • Number of test queries varied for different topics to obtain the required (0.95) statistical significance

  28. Baseline Methods • Used four baseline methods for comparison • CORI (A standard Collection Selection approach) • GoogleBase • All Sources • Only on the crawled sources • USR: Undifferentiated Source Rank • One rank per source, independent of query topic • DSR: Domain Specific Source Rank • Assumes oracular information on the topic of the source as well as query

  29. Source selection • Agreement based selection models (TSR, USR and DSR) use a weighted combination of importance and relevance scores • Example: TSR(0.1) represents 0.9xCORI + 0.1xTSR • For each query q, top-k sources are selected • Google Base is made to query only on these to top-k sources

  30. Tuple ranking and relevance evaluation • Google Base’s tuple ranking is used for ranking resulting tuples • Top-5 results returned were manually classified as relevant or irrelevant • Result classification was rule based

  31. TSR precision exceeds that of similarity-based measures by 85% Comparison of top-5 precision of TSR(0.1) and query similarity based methods: CORI and Google Base

  32. TSR significantly out-performs all query-similarity based measures for all topics Comparison of topic-wise top-5 precision of TSR(0.1) and query similarity based methods: CORI and Google Base

  33. TSR precision exceeds USR(0.1) by 18% and USR(1.0) by 40% Comparison of top-5 precision of TSR(0.1) and agreement based methods: USR(0.1) and USR(1.0)

  34. For three out of the four topics, TSR(0.1) out-performs USR(0.1) and USR(1.0) with confidence levels 0.95 or more Comparison of topic-wise top-5 precision of TSR(0.1) and agreement based methods: USR(0.1) and USR(1.0)

  35. TSR(0.1) is able to match DSR(0.1)’s performance Comparison of top-5 precision of TSR(0.1) and oracular DSR(0.1)

  36. TSR(0.1) matches DSR(0.1) performance across all topics indicating its effectiveness in identifying important sources across all topics Comparison of topic-wise top-5 precision of TSR(0.1) and oracular DSR(0.1)

  37. Conclusion • Attempted multi-topic source selection sensitive to trustworthiness and importance for the deep-web • Introduced topic-sensitive SourceRank (TSR) • Our experiments on more than a thousand deep-web sources show that a TSR-based approach is highly effective in extending SourceRank to multi-topic deep-web

  38. Conclusion contd. • TSR out-performs query-similarity based measures by around 85% in precision • TSR results in statistically significant precision improvements over other baseline agreement-based methods • Comparison with oracular DSR approach reveals effectiveness of TSR for topic-specific query and source classification and subsequent source selection

More Related